Bayesian testing of an exponential point null hypothesis

摘要: The problem of testing a point null hypothesis involving an exponential mean is usual interpretation P-values as evidence against precise hypotheses faulty. As in Berger and Delampady (1986) Sellke (1987), lower bounds on Bayesian measures over wide classes priors are found emphasizing the conflict between posterior probabilities P-values. A hierarchical Bayes approach also considered alternative to computing “automatic” significance tests which further illustrates that highly misleading for hypotheses.